Robot-assisted Transcranial Magnetic Stimulation (Robo-TMS): A Review

Wenzhi Bai, Andrew Weightman, Rory J. O’Connor, Zhengtao Ding, Mingming Zhang, Sheng Quan Xie, Zhenhong Li · 2025

Read Robot-assisted Transcranial Magnetic Stimulation (Robo-TMS): A Review by Wenzhi Bai, Andrew Weightman, Rory J. O’Connor, Zhengtao Ding, Mingming Zhang, Sheng Quan Xie, Zhenhong Li (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-04345.txt.

Introduction

Wenzhi Bai and colleagues review robot-assisted transcranial magnetic stimulation, or Robo-TMS, from the view of how it is built. It is a safe brain treatment that does not cut the skin. Its use in care and research grows. To hold a coil over one spot for a long time is hard for a human hand. A robot can hold it still. But no review of the build side had been made.

The team studies four parts: hardware and how parts are joined, calibration and registration, neuronavigation, and control. Wider use in care is held back by three things. Gain for patients is not yet proven, the work is hard to run, and cost is large. Good paths include tracking with no markers, non-rigid registration, and self-guided calibration. They also include field models that learn, MRI made for each person, and care at more than one spot.

Catalog record

CREATORWenzhi Bai, Andrew Weightman, Rory J. O’Connor, Zhengtao Ding, Mingming Zhang, Sheng Quan Xie, Zhenhong Li
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE41 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-04345

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Robot-assisted Transcranial Magnetic Stimulation (Robo-TMS): A Review Thanks: IEEE Transactions on Neural Systems and Rehabilitation Engineering. Thanks: DOI: 10.1109/TNSRE.2025.3585651 Thanks: Corresponding Author (email:[email protected]) Wenzhi Bai, Andrew Weightman, Rory J. O’Connor, Zhengtao Ding, Mingming Zhang, Sheng Quan Xie, Zhenhong Li Abstract Transcranial magnetic stimulation (TMS) is a non-invasive and safe brain stimulation procedure with growing applications in clinical treatments and neuroscience research. However, achieving precise stimulation over prolonged sessions poses significant challenges. By integrating advanced robotics with conventional TMS , robot-assisted TMS (Robo-TMS) has emerged as a promising solution to enhance efficacy and streamline procedures. Despite growing interest, a comprehensive review from an engineering perspective has been notably absent. This paper systematically examines four critical aspects of Robo-TMS : hardware and integration; calibration and registration; neuronavigation systems; and control systems. We review state-of-the-art technologies in each area, identify current limitations, and propose future research directions. Our findings suggest that broader clinical adoption of Robo-TMS is currently limited by unverified clinical applicability, high operational complexity, and substantial implementation costs. Emerging technologies—including marker-less tracking, non-rigid registration, learning-based electric field (E-field) modelling, individualised magnetic resonance imaging (MRI) generation, robot-assisted multi-locus TMS ( Robo-mTMS ), and automated calibration and registration—present promising pathways to address these challenges. I Introduction Transcranial magnetic stimulation ( TMS ) is a non-invasive and safe brain stimulation procedure first introduced in 1985 [ 1 ] . It utilises a time-varying electromagnetic field generated by coils to induce an electric current in targeted brain regions [ 2 , 3 , 4 ] . As neurogenic diseases and mental disorders remain leading causes of disability and continue to pose a significant global health burden [ 5 ] , there is a growing demand for effective diagnostic and therapeutic tools. Leveraging its neurostimulation and neuromodulation capabilities, TMS has been extensively explored for diagnosing and treating various neurogenic diseases and mental disorders [ 6 , 7 , 8 ] , as well as for brain mapping research [ 9 , 10 , 11 , 12 ] . Since TMS does not require active subject participation, it can be effectively performed on individuals who are paralysed, sedated, or uncooperative [ 13 ] . Following its initial approval by the U.S. Food and Drug Administration (FDA) in 2008, TMS has seen growing adoption in neurosurgery planning and treating depression, obsessive-compulsive disorder (OCD) , migraines, and smoking cessation especially when standard treatments have proven ineffective [ 14 , 15 ] . Fig. 1: A typical application scenario of Robo-TMS . The subject lies in a chair with a head marker attached. An optical tracking camera monitors head and coil markers to guide the robot arm in compensating for head movements, maintaining a consistent coil-to-head transformation for accurate stimulation. The TMS stimulator generates current pulses into the TMS coil, and the resulting induced E-field within the subject’s head is visualised by the neuronavigation software on the monitor. In conventional TMS , clinicians cannot directly visualise the stimulation spot beneath the scalp, and even the corresponding location on the subject’s head is often obscured by the stimulation coil. Maintaining precise targeting throughout prolonged sessions is further complicated by the need to manually hold the coil. To address these challenges, advanced robotics innovations have been integrated into conventional TMS to enhance clinical outcomes. These advancements include neuronavigation systems for accurate stimulation targeting, E-field modelling for the estimation of induced E-field distribution, optical tracking systems for subject’s head movements compensation, and robot arms for stable and precise coil placement. Reflecting these new capabilities, the terminology used to describe these procedures has evolved: TMS enhanced with neuronavigation systems is often referred to as navigated TMS (nTMS) [ 16 ] ; automated systems that incorporate robot arms are termed robotic TMS , robotised TMS , robot-assisted TMS (RA-TMS) , or robot-assisted image-guided TMS (RI-TMS) [ 17 ] . To ensure clarity and consistency, we adopt the term robot-assisted TMS (Robo-TMS) to describe the integrated system illustrated in Fig. 1 . Recent literature on PubMed indicates a growing interest among medical researchers in more advanced TMS devices to meet evolving clinical needs. However, comprehensive reviews on Robo-TMS , particularly from an engineering perspective, are absent. Existing reviews primarily focus on the fundamental aspects of TMS [ 18 , 13 , 3 ] and its clinical applications [ 9 , 7 , 8 ] . Some reviews emphasise the importance of neuronavigation systems and the necessity for high accuracy [ 19 , 20 , 21 ] , while others delve into specific aspects like E-field modelling [ 22 , 23 ] , brain mapping [ 24 ] , and accuracy comparisons across different systems [ 25 , 26 , 27 , 28 ] . Although early insights into Robo-TMS are provided by a book [ 29 ] and a doctoral thesis [ 30 ] over a decade ago, the rapid advancements in robotics underscore the need for a comprehensive review that introduces emerging engineering tools for addressing current challenges in TMS practice and informs the development of technology that aligns with clinical requirements. This review focuses on the core technologies and components of Robo-TMS , while omitting elements that have been widely discussed in conventional TMS , such as pulse waveform, cooling, power supply and etc. The main objectives are 1. to identify the status and challenges in four critical aspects: hardware and integration, calibration and registration, neuronavigation systems, and control systems; 2. to bridge the gap between the medical and engineering perspectives, providing a comprehensive understanding of Robo-TMS ; 3. to propose directions for further research and development by discussing how to potentially enhance clinical outcomes and reduce operational costs. To the best of our knowledge, this is the first review to comprehensively survey the status and challenges of Robo-TMS from an engineering perspective. We hope it can foster collaboration among clinicians, engineers and researchers, ultimately driving future advancements in the field. TABLE I: The Comparison of Typical TMS Systems Conventional TMS System Industrial Robot-based Robo-TMS System Specialised Robot-based Robo-TMS System Holders Passive Mechanical Holder Active Industrial Robot Arm Active Specialised Robot Arm Force / Torque Sensors None Required Required Optical Tracking Optional Required Required E-field Modelling Optional Recommended Recommended Accuracy * \sim 6\mathrm{mm}/\sim 3^{\circ} [ 31 , 32 , 33 ] \sim 2\mathrm{mm}/\sim 1.5^{\circ} [ 31 , 32 , 33 ] \sim 2\mathrm{mm}/\sim 1.5^{\circ} [ 31 , 32 , 33 ] Contact Force * 6-10\mathrm{N} [ 34 ] \sim 2.5\mathrm{N} [ 35 , 32 ] \sim 2.5\mathrm{N} [ 35 , 32 ] Examples Magstim TM Horizon Lite [ 36 ] Axilum Robotics TM TMS-Cobot [ 37 ] Yiruide TM Mag-aim [ 38 ] * Accuracy and contact force listed in the table are for specific commercial systems and experimental prototypes. II Hardware and Integration II-A System Setup Manually holding the coil for prolonged TMS procedures not only exhausts clinicians but also leads to inaccuracies caused by fatigue, affecting stimulation efficacy. While mechanical holders offer some stability [ 39 ] , they are unable to compensate for unexpected head movements. Compromised solutions, such as instructing subjects to remain still or employing head restraints, often cause discomfort and stress. Robo-TMS , which integrates neuronavigation with a robot arm to ensure accurate coil targeting and precise coil

Read the complete text